Torch7: no broadcasting in torch?

[yangky11]

Hi all,

Is there any mechanism in Torch similar to broadcasting in numpy?

The problem is that I have a NxD matrix X and a vector y of length N. I would like to add each element in y to corresponding row of X. If it were numpy, then the code would be:

X + y[:, numpy.newaxis]

Is there an elegant way to do it in torch?

Thanks.

[nicholas-leonard]

@yangky11

X = torch.randn(3,4)
y = torch.randn(3)
z = X + y:view(3,1):expandAs(X)